AI Chatbots Remain Confident — Even When They’re Wrong
Artificial intelligence chatbots are everywhere these days, integrated into decision-making in domains like health, law, education and customer support. Confident but incorrect answers can mislead users, especially when the AI sounds authoritative. Research from Carnegie Mellon University explains what happens when these handy tools overestimate their own abilities and how humans can build trust.
Researchers asked both human participants and four large language models (LLMs) how confident they felt in their ability to answer trivia questions, predict the outcomes of NFL games or Academy Award ceremonies, or play a Pictionary-like image identification game. Both the people and the LLMs tended to be overconfident about how they would hypothetically perform. Interestingly, they also answered questions or identified images with relatively similar success rates.
However, when the participants and LLMs were asked retroactively how well they thought they did, only the humans appeared able to adjust expectations, according to a study published in the journal Memory & Cognition(opens in new window) .
